AMIVA-F

Appropriate functioning of Filamin C (FlnC) is crucial for structural integrity and cell signalling of the sarcomere and is therefore required for proper heart muscle function. High-throughput screening in cardiomyopathy cohorts determined a prominent role for FlnC in both hyptertrophic, as well as dilated cardiomyopaties [1].

A common feature shared among those patients suffering form both inherited cardiac diseases is the prevalence of single point missense variants in FlnC, leading to a single amino acid substitution in the FlnC protein. Among the potential variants of FlnC, some are associated with a malignant clinical course and a high risk of sudden cardiac death, while others are perfectly tolerated.
AMIVA-F is a machine learning based AI that is capable of predicting the effect of single point mutations in FlnC while also providing molecular insights into its predictions. It is trained on a dataset consisting of clinically and experimentally verified disease as well as neutral related mutations.
Based on structural and biophyscial parameters, AMIVA-F achieves similar accuracy as commonly used classifiers (Balanced AUC-ROC, 94%).

While agreeing on many predictions with commonly used predictors like AlphaMissense [2] , AMIVA-F significantly improves sampling of "uncertainly classified" mutations with the additional benefit of explainable predictions. Overall, combining multiple predictors together improves further classification accuracy and it is suggested to additionally utilize other existing tools for predictions.
If pathogenic variants are recognized early, pre-emptive medical treatment can prevent sudden cardiac death and allow for an otherwise healthy normal life.

1. Verdonschot JAJ, Vanhoutte EK, Claes GRF, et al. A mutation update for the FLNC gene in myopathies and cardiomyopathies. Hum Mutat. 2020;41(6):1091-1111. doi:10.1002/humu.24004
2. Jun Cheng et al., Accurate proteome-wide missense variant effect prediction with AlphaMissense.Science381,eadg7492(2023).DOI:10.1126/science.adg7492